Sinusitis is the 5th most common diagnosis warranting an antimicrobial prescription in the US. However, little information is available regarding which signs, symptoms, or bacteriologic findings best define a diagnosis of sinusitis in children. Without clear clinical diagnostic criteria, practitioners have had difficulty differentiating children with acute bacterial sinusitis from children with simple colds. The proposed research will examine the diagnostic value of clinical and bacteriologic findings, individually and combined into a clinical prediction tool, for acute bacterial sinusitis. Use of these criteria has the potential to reduce inappropriate antimicrobial use for this frequently occurring condition. Furthermore, there are no adequately validated methods of measuring treatment response in sinusitis studies. Since resolution of symptoms is an important goal of antimicrobial therapy, a symptom scale would logically be a desirable method of measuring outcome in clinical trials. Development of a valid symptom scale, which is the second aim of this proposal, will allow researchers to compare the effectiveness of various treatment strategies in children with acute sinusitis. We will enroll a prospective cohort of 260 children aged 2-12 years with suspected sinusitis. At baseline, we will collect detailed information regarding signs and symptoms, along with a culture from the nasopharynx. All children will undergo sinus radiography;those with radiographic evidence of sinusitis will be treated with high- dose amoxicillin-clavulanate and followed for 10 days using a preliminary symptom scale. We will use this information to determine which findings best identify children with acute sinusitis as defined by positive radiographs and to validate a symptom scale for use in future trials. This preliminary work will lay the necessary groundwork for a definitive multicenter randomized placebo- controlled study evaluating the efficacy of antimicrobial therapy in children with acute bacterial sinusitis. The clinical prediction rule will inform the inclusion criteria for the study and the symptom scale will be used as the primary outcome measure of outcome. PUBLIC HEALTH RELEVANCE: Using criteria identified through this study, practitioners will be able to more accurately separate children with acute sinusitis who need antibiotics from children with colds who do not. In addition, we will also develop a symptom scale which will allow researchers to compare the effectiveness of different treatment strategies for children with acute bacterial sinusitis.